Contributors Guide#

Interested in helping build MetPy? Have code from your research that you believe others will find useful? Have a few minutes to tackle an issue? In this guide we will get you setup and integrated into contributing to MetPy!


First off, thank you for considering contributing to MetPy. MetPy is community-driven project. It’s people like you that make MetPy useful and successful. There are many ways to contribute, from writing tutorials or examples, improvements to the documentation, submitting bug reports and feature requests, or even writing code which can be incorporated into MetPy for everyone to use.

Following these guidelines helps to communicate that you respect the time of the developers managing and developing this open source project. In return, they should reciprocate that respect in addressing your issue, assessing changes, and helping you finalize your pull requests.

So, please take a few minutes to read through this guide and get setup for success with your MetPy contributions. We’re glad you’re here!

What Can I Do?#

  • Tackle any issues you wish! We have a special label for issues that beginners might want to try. Have a look at our current beginner issues. Also have a look at if the issue is already assigned to someone - this helps us make sure that work is not duplicated if the issue is already being worked on by Unidata Staff.

  • Contribute code you already have. It does not need to be perfect! We will help you clean things up, test it, etc.

  • Make a tutorial or example of how to do something.

  • Improve documentation of a feature you found troublesome.

  • File a new issue if you run into problems!

Ground Rules#

The goal is to maintain a diverse community that’s pleasant for everyone. Please be considerate and respectful of others by following our code of conduct.

Other items:

  • Each pull request should consist of a logical collection of changes. You can include multiple bug fixes in a single pull request, but they should be related. For unrelated changes, please submit multiple pull requests.

  • Do not commit changes to files that are irrelevant to your feature or bug fix (eg: .gitignore).

  • Be willing to accept criticism and work on improving your code; we don’t want to break other users’ code, so care must be taken not to introduce bugs.

  • Be aware that the pull request review process is not immediate, and is generally proportional to the size of the pull request.

  • Function arguments:

    • Use full names for parameters rather than symbols (e.g. temperature instead of t)

    • Order: pressure/height -> temperature/wind -> moisture (in general, but not a hard and fast rule–like to allow for some default arguments).

Reporting a bug#

The easiest way to get involved is to report issues you encounter when using MetPy or by requesting something you think is missing.

  • Head over to the issues page.

  • Search to see if your issue already exists or has even been solved previously.

  • If you indeed have a new issue or request, click the “New Issue” button.

  • Fill in as much of the issue template as is relevant. Please be as specific as possible. Include the version of the code you were using, as well as what operating system you are running. If possible, include complete, minimal example code that reproduces the problem.

Setting up your development environment#

We recommend using the conda package manager for your Python environments. This requires some comfort with the command line and a little git knowledge. Our recommended setup for contributing:

Install miniconda on your system. You may have to restart your prompt for the remaining steps to work.

Install git (link with instructions) on your system if not already available (check with git --version at the command line.) This can also be installed from a variety of package managers, including conda if needed.

Login to your GitHub account and make a fork of the MetPy repository by clicking the “Fork” button. Clone your fork of the MetPy repository (in terminal on Mac/Linux or git shell/GUI on Windows) to the location you’d like to keep it. We are partial to creating a git_repos or projects directory in our home folder.

git clone<your-user-name>/metpy.git

Navigate to that folder in the terminal or in Anaconda Prompt if you’re on Windows. The remainder of the instructions will take place within this directory.

cd metpy

Connect your repository to the upstream (main project).

git remote add unidata

Create a new conda environment for us to configure, and give it a name. After -n you can specify any name you’d like; here we’ve chosen devel.

conda create -n devel

IMPORTANT: Always activate this environment when developing and testing your changes!

conda activate devel

You will have to do this any time you re-open your prompt. Currently there are no packages in this environment, let’s change that. Configure this environment so that we can reach conda-forge for the specific packages we depend on.

conda config --env --add channels conda-forge --add channels conda-forge/label/testing

Install the necessary dependency packages from conda-forge. Remember that these must be executed within the metpy directory.

conda install --file ci/requirements.txt --file ci/extra_requirements.txt --file ci/test_requirements.txt

Finally, create an editable install of MetPy that will update with your development!

pip install -e .

Note sections on documentation and code style below, where you may need to install a few more packages into your new environment.

Now you’re all set! You have an environment called devel that you can work in. Remember, you will need to activate this environment the next time you want to use it after closing the terminal. If you want to get back to the root environment, run conda deactivate.

Pull Requests#

The changes to the MetPy source (and documentation) should be made via GitHub pull requests against main, even for those with administration rights. While it’s tempting to make changes directly to main and push them up, it is better to make a pull request so that others can give feedback. If nothing else, this gives a chance for the automated tests to run on the PR. This can eliminate “brown paper bag” moments with buggy commits on the main branch.

During the Pull Request process, before the final merge, it’s a good idea to rebase the branch and squash together smaller commits. It’s not necessary to flatten the entire branch, but it can be nice to eliminate small fixes and get the merge down to logically arranged commits. This can also be used to hide sins from history–this is the only chance, since once it hits main, it’s there forever!

Working on your first Pull Request? You can learn how from this free video series How to Contribute to an Open Source Project on GitHub, Aaron Meurer’s tutorial on the git workflow, or the guide “How to Contribute to Open Source”.

Commit the changes you made. Chris Beams has written a guide on how to write good commit messages.

Push to your fork and submit a pull request. For the Pull Request to be accepted, you need to agree to the MetPy Contributor License Agreement (CLA). This will be handled automatically upon submission of a Pull Request. See here for more explanation and rationale behind MetPy’s CLA.

Source Code#

MetPy’s source code is located in the src/ directory in the root of the repository. Within src/ is the metpy/ directory, which is the base package. Inside here are the main top-level subpackages of MetPy:

  • calc: Calculations and tools

  • interpolate: Interpolating data points to other locations

  • io: Tools for reading and writing files

  • plots: Plotting tools using Matplotlib (and Cartopy)


Now that you’ve made your awesome contribution, it’s time to tell the world how to use it. Writing documentation strings is really important to make sure others use your functionality properly. Didn’t write new functions? That’s fine, but be sure that the documentation for the code you touched is still in great shape. It is not uncommon to find some strange wording or clarification that you can take care of while you are here. If you added a new function make sure that it gets marked as included if appropriate in the GEMPAK conversion table.

You can write examples in the documentation if they are simple concepts to demonstrate. If your feature is more complex, consider adding to the examples or tutorials for MetPy.

You can build the documentation locally to see how your changes will look. After setting up your development environment above, from within the metpy directory with your devel environment active, use conda install --file ci/doc_requirements.txt to install required packages to build our documentation. Then, still from within your devel environment,

  • Navigate to the docs folder cd docs

  • Remove any old builds and build the current docs make clean html

    • (Try make cleanall html if make clean html fails)

  • Open docs/build/html/index.html and see your changes!

The MetPy documentation relies on the Pydata Sphinx Theme for style and functionality. The theme includes some live javascript elements, including the version switcher. To test these elements, use our to deploy the built docs html files to a local server. If testing changes to pst-versions.json locally, change html_theme_options['switcher']['json_url'] to reference /MetPy/pst-versions.json and the builds should pass and reflect any testing changes to docs/test-server/pst-versions.json. Documentation builds may fail if the links in the json fail to resolve. Change html_theme_options['check_switcher'] to False in to bypass this behavior. Note: for production, pst-versions.json must live and is automatically updated on the online MetPy documentation via the gh-pages branch on GitHub.


Unit tests are the lifeblood of the project, as it ensures that we can continue to add and change the code and stay confident that things have not broken. Running the tests requires pytest, which is easily available through conda or pip. It was also installed if you made our default devel environment.

Running Tests#

Running the tests can be done by running pytest

Running the whole test suite isn’t that slow, but can be a burden if you’re working on just one module or a specific test. It is easy to run tests on a single directory:

pytest tests/calc

A specific test can be run as:

pytest -k test_my_test_func_name

Writing Tests#

Tests should ideally hit all of the lines of code added or changed. We have automated services that can help track down lines of code that are missed by tests. Watching the coverage has even helped find sections of dead code that could be removed!

Let’s say we are adding a simple function to add two numbers and return the result as a float or as a string. (This would be a silly function, but go with us here for demonstration purposes.)

def add_as_float_or_string(a, b, as_string=False):
    res = a + b
    if as_string:
       return string(res)
    return res

I can see two easy tests here: one for the results as a float and one for the results as a string. If I had added this to the calc module, I’d add those two tests in tests/calc/

def test_add_as_float_or_string_defaults():
    res = add_as_float_or_string(3, 4)
    assert(res, 7)

def test_add_as_float_or_string_string_return():
    res = add_as_float_or_string(3, 4, as_string=True)
    assert(res, '7')

There are plenty of more advanced testing concepts, like dealing with floating point comparisons, parameterizing tests, testing that exceptions are raised, and more. Have a look at the existing tests to get an idea of some of the common patterns.

Image tests#

Some tests (for matplotlib plotting code) are done as an image comparison, using the pytest-mpl plugin. By following the guide above, you should have a testing install of matplotlib, which will guarantee that your image tests behave exactly the same as ours. To run these tests, use:

pytest --mpl

When adding new image comparison tests, start by creating the baseline images for the tests:

pytest --mpl-generate-path=baseline

That command runs the tests and saves the images in the baseline directory. For MetPy this is generally tests/plots/baseline/. We recommend using the -k flag to run only the test you just created for this step.

For more information, see the docs for pytest-mpl.

Cached Data Files#

MetPy keeps some test data, as well as things like shape files for US counties in a data cache supported by the pooch library. To add files to this, please ensure they are as small as possible. Put the files in the staticdata directory. Then run this command in the root of the MetPy repository to recreate the data registry:

python -c "import pooch; pooch.make_registry('staticdata', 'src/metpy/static-data-manifest.txt')"

Make sure that no system files (like .DS_Store) are in the manifest and add it to your contribution.

Code Style#

MetPy uses the Python code style outlined in PEP8. For better or worse, this is what the majority of the Python world uses. The one deviation is that line length limit is 95 characters. 80 is a good target, but some times longer lines are needed.

While the authors are no fans of blind adherence to style and so-called project “clean-ups” that go through and correct code style, MetPy has adopted this style from the outset. Therefore, it makes sense to enforce this style as code is added to keep everything clean and uniform. To this end, part of the automated testing for MetPy checks style. To check style locally within the source directory you can use the ruff and flake8 tools. After setting up your development environment above, from within the metpy directory with your devel environment active, install the code style tools we use with conda install --file ci/linting_requirements.txt. Checking your code style is then as easy as running ruff check . ; flake8 . in the base of the repository.

You can also just submit your PR and the kind robots will comment on all style violations as well. It can be a pain to make sure you have the right number of spaces around things, imports in order, and all of the other nits that the bots will find. It is very important though as this consistent style helps us keep MetPy readable, maintainable, and uniform.

What happens after the pull request#

You’ve make your changes, documented them, added some tests, and submitted a pull request. What now?

Automated Testing#

First, our army of never sleeping robots will begin a series of automated checks. The test suite, documentation, style, and more will be checked on various versions of Python with current and legacy packages. Travis CI and GitHub Actions will run testing on Linux, and Mac, and Windows. Other services will kick in and check if there is a drop in code coverage or any style variations that should be corrected. If you see a red mark by a service, something failed and clicking the “Details” link will give you more information. We’re happy to help if you are stuck.

The robots can be difficult to satisfy, but they are there to help everyone write better code. In some cases, there will be exceptions to their suggestions, but these are rare. If you make changes to your code and push again, the tests will automatically run again.

Code Review#

At this point you’re waiting on us. You should expect to hear at least a comment within a couple of days. We may suggest some changes or improvements or alternatives.

Some things that will increase the chance that your pull request is accepted quickly:

Pull requests will automatically have tests run by Travis. This includes running both the unit tests as well as the flake8 code linter.


Once we’re all happy with the pull request, it’s time for it to get merged in. Only the maintainers can merge pull requests and you should never merge a pull request you have commits on as it circumvents the code review. If this is your first or second pull request, we’ll likely help by rebasing and cleaning up the commit history for you. As your development skills increase, we’ll help you learn how to do this.

More Questions?#

If you’re stuck somewhere or are interested in being a part of the community in other ways, feel free to contact us:

Further Reading#

There are a ton of great resources out there on contributing to open source and on the importance of writing tested and maintainable software.